Finding drugs that target breast cancer invasion (the first step in metastasis) without affecting normal tissue is an important unmet goal. One protein target of high current interest is the molecular chaperone Hsp90, which functions in cancer progression and metastasis as well as normal cell function. Hsp90 inhibitors such as ganetespib are currently in clinical trials, however they may have serious side effects based on the many proteins that require Hsp90 for activation. We introduced the concept of inhibiting Hsp90 outside of cancer cells to circumvent this problem. My lab first showed that extracellular Hsp90 (eHsp90) is released by cancer cells and that it acts in cancer invasion by activation of Matrix Metalloproteinase-2. Since then, we and others have shown that eHsp90 activates many pro-proteins in the extracellular media to enhance invasion including Lysyl Oxidase-like 2 (LOXL2), which remodels the extracellular matrix and is well implicated in cancer. We also showed that eHsp90 is released from cells via exosomes and, in new data, we show that inhibiting Hsp90 reduces exosome release and uptake, which has been implicated in tumor communication during invasion. Inhibiting eHsp90 may prevent this thus reducing invasion and metastasis. Our long-term goal is to achieve a clinical trial for an eHsp90 inhibitor for breast cancer. Recently we showed that STA-12-7191 (an impermeant derivative of ganetespib) inhibits cancer cell motility but is 5-fold less toxic to normal cells. These findings suggest that specifically inhibiting eHsp90 will benefit cancer treatment by preventing activation of LOXL2 and exosome release, reducing invasion without the damaging effects of inhibition of Hsp90's intracellular functions. We will test this hypothesis in the following Aims: 1. Test if LOXL2 is a bona fide eHsp90 client thereby providing a biomarker for eHsp90 inhibition in vivo; 2. Determine how eHsp90 acts in exosome trafficking and show this is pro-invasive; 3. Test the association between serum eHsp90 levels and metastatic breast cancer using clinically annotated patient sera and a novel ELISA diagnostic; and 4. Test the role of eHsp90 in breast cancer metastasis in human breast-to-bone xenograft model. We bring together an outstanding collaborative team including the PI who pioneered eHsp90, Dr. Ying who developed ganetespib and STA-12-7191 and experts in Hsp90 (Neckers), breast cancer biomarkers (Seewaldt and Luo) and metastasis animal models (Kuperwasser). This study is significant because it would validate eHsp90 as an important drug target for treating metastatic breast cancer and implicate it in two important pro-invasive processes: exosome-based tumor communication and ECM remodeling by LOXL2. Importantly, this study could lead to a drug that would inhibit these processes with fewer side effects than the Hsp90 inhibitors currently in human trials. In addition, it would impact other cancers and other diseases given eHsp90's multiple roles in pathogenesis. The work is innovative because it tests a novel mechanism for exosome trafficking and introduces a novel compound to inhibit eHsp90 and testing its benefit in cancer.